“Lifetime commissions on returning customers” sounds like a forecast. It is a description of a program’s attribution promise, subject to its current terms. To make a campaign decision, the affiliate still needs to answer three separate questions: will customers reorder, will those orders be attributed and finalized, and will the resulting commissions exceed the cost of acquiring them?
Those questions matter because customer lifetime value and affiliate revenue are different measures. A seller may earn from a customer over time. An affiliate can model only the commissions that are attributed under the program rules and ultimately become payable.
What NutriProfits currently says about repeat orders
NutriProfits’ public FAQ says that returning customers remain assigned to the affiliate account and that repeat purchases may earn up to 20% commission. It also describes lifetime cookie tracking. Treat these as the program’s own current statements, not as a guarantee of customer retention or future income. Offer-level rates and conditions should be checked in the account before publication.
The network’s Terms of Use add details that affect the math: the fee is shown in the administration panel and may change; a transaction qualifies after it is fully paid and shipped; and fees can be deducted when a customer returns a product or files a payment dispute. The terms also describe monthly payment of finalized transactions from the prior month. That means an attributed order, a finalized commission, and cash received can be different events on different dates.
A cookie or account attribution rule concerns credit assignment. It does not make an order happen, guarantee a repeat rate, prove the same rate applies to every offer, or remove the effects of cancellations, returns, payment disputes, and current program terms.
Separate the model into four numbers
For each acquisition cohort, track these quantities separately:
- Acquisition cost per first customer: traffic spend plus a reasonable allocation of content, production, tools, and operating costs.
- Initial commission: the commission attached to the first order after its status is understood.
- Repeat behavior: the share of first-time customers with a later order and the number of repeat orders per customer over a defined period.
- Finalized payable commission: the amount left after cancellations, returns, disputes, and the program’s own qualification rules.
A useful planning equation is:
Expected commission per acquired customer = expected finalized initial commission + (repeat-order probability × expected repeat orders per returning customer × finalized commission per repeat order).
Expected contribution per acquired customer = expected commission per acquired customer − acquisition cost per first customer.
These are estimates until they are measured. If the network does not provide enough reporting to distinguish initial from repeat orders for a cohort, do not fill that gap with an assumed repeat rate. Use the one-order case for the bid decision and mark the unavailable data as unknown.
A hypothetical sensitivity check
The following numbers are invented solely to show how the equation works. They are not NutriProfits benchmarks or a prediction for any product. Assume an initial finalized commission of $21, a repeat commission of $12, and an acquisition cost of $22 per first customer. The example ignores taxes and assumes the commissions shown are finalized after any reversals.
| Hypothetical case | Share with a repeat order | Average repeat orders per returning customer | Expected repeat commission per acquired customer | Expected total commission | Contribution before tax |
|---|---|---|---|---|---|
| No repeat observed | 0% | 0 | $0.00 | $21.00 | −$1.00 |
| Modest repeat assumption | 30% | 1 | $3.60 | $24.60 | +$2.60 |
| Higher repeat assumption | 45% | 2 | $10.80 | $31.80 | +$9.80 |
The calculation is straightforward: in the middle case, 0.30 × 1 × $12 = $3.60 in expected repeat commission per acquired customer. Add the $21 initial commission, then subtract the $22 acquisition cost. The result is positive only under that assumed repeat behavior.
The table shows sensitivity, not evidence that either repeat rate is normal. Before data exists, use scenarios to see which assumptions drive the decision. Do not choose the most favorable scenario simply because it supports a higher bid. Once cohort data becomes available, replace assumptions with observed values and keep the sample size and observation period visible.
Measure repeat behavior without overclaiming
NutriProfits explains in its FAQ that a Sub ID can identify the effectiveness of a particular link. Use separate identifiers for meaningful comparisons, such as article, angle, or market, and keep a simple record of what each identifier means. Avoid putting a reader’s name, email, or other personal information into a tracking value.
A Sub ID can help organize attribution data; it does not automatically provide a customer-level retention report. Confirm what the account dashboard actually exposes. To evaluate repeat economics, you would ideally need counts of first orders, later orders attributed to the same cohort, order status, commission amount, and payment date. If the report gives only clicks and total transactions, it cannot establish an individual customer’s repeat pattern.
Set the observation window from the product’s documented consumption information, fulfillment timing, return conditions, and reporting delay. There is no single period that works for every product or market. Compare cohorts over equivalent windows and show the number of first-time customers behind each rate. A small cohort can swing sharply when one or two orders change; label it preliminary and avoid treating it as a stable forecast.
A returning order is useful evidence of repeat behavior. It is not, on its own, proof that the affiliate caused retention or that the same rate will continue. Price changes, availability, shipping, product experience, the seller’s communications, and customer circumstances may all affect the result.
What the affiliate can influence
An affiliate does not control product quality, fulfillment, stock, price, post-purchase service, retention emails, or the network’s attribution rules. The work that is within reach is narrower:
- Audience fit: select a product and article topic that serve the same reader need.
- Expectation setting: describe documented features, terms, and limits accurately. Do not promise a health or beauty outcome that the available evidence cannot support.
- Useful content: help readers compare options or understand a purchase decision, even if they leave without using an affiliate link.
- Measurement discipline: preserve campaign definitions, dates, source reports, and order-status differences so later comparisons remain interpretable.
If a recommendation includes an affiliate link, readers should understand the commercial relationship. For U.S. endorsements, the Federal Trade Commission’s affiliate marketing guidance says a commission relationship should be disclosed clearly and conspicuously, close enough to the recommendation for readers to notice it. Other markets may apply different rules.
A conservative bidding rule
When a campaign has no reliable repeat data, decide whether the initial order can support the acquisition cost on its own. If it cannot, record the size of the gap and what evidence would be needed to test a repeat-based case. Do not treat a program’s attribution promise as revenue already earned.
When repeat reporting is available, add expected repeat commission gradually and only after checking that it is finalized, attributed to the intended cohort, and observed over a comparable period. Keep the one-order model beside the repeat model so the team can see how much the decision depends on future purchases. If the campaign works only under an unverified repeat assumption, the campaign is still an experiment, not a proven unit-economics case.
A generous attribution window does not justify spending commission that has not been earned. Keep the acquisition budget tied to the value the evidence can support, and let finalized repeat purchases earn their place in the model. The difference between those two numbers tells the publisher how much of the business case still depends on customer behavior it has yet to observe.
